ViewVC Help

View File | Revision Log | Show Annotations | Root Listing

root/cvsroot/COMP/CRAB/python/Splitter.py

Comparing COMP/CRAB/python/Splitter.py (file contents):
Revision 1.23 by spiga, Fri Jun 19 09:54:22 2009 UTC vs.
Revision 1.40 by spiga, Mon Jun 7 10:38:39 2010 UTC

#	Line 1 | Line 1
1	+
2	+	__revision__ = "$Id$"
3	+	__version__ = "$Revision$"
4	+
5		import common
6		from crab_exceptions import *
7		from crab_util import *
8	+
9	+	from WMCore.DataStructs.File import File
10	+	from WMCore.DataStructs.Fileset import Fileset
11	+	from WMCore.DataStructs.Run import Run
12	+	from WMCore.DataStructs.Subscription import Subscription
13	+	from WMCore.DataStructs.Workflow import Workflow
14	+	from WMCore.JobSplitting.SplitterFactory import SplitterFactory
15		from WMCore.SiteScreening.BlackWhiteListParser import SEBlackWhiteListParser
16	+	from LumiList import LumiList
17
18		class JobSplitter:
19		def __init__( self, cfg_params,  args ):
20		self.cfg_params = cfg_params
21		self.args=args
22	+
23	+	self.lumisPerJob = -1
24	+	self.totalNLumis = 0
25	+	self.theNumberOfJobs = 0
26	+	self.limitNJobs = False
27	+	self.limitTotalLumis = False
28	+	self.limitJobLumis = False
29	+
30		#self.maxEvents
31		# init BlackWhiteListParser
32	<	seWhiteList = cfg_params.get('GRID.se_white_list',[])
32	>	self.seWhiteList = cfg_params.get('GRID.se_white_list',[])
33		seBlackList = cfg_params.get('GRID.se_black_list',[])
34	<	self.blackWhiteListParser = SEBlackWhiteListParser(seWhiteList, seBlackList, common.logger())
34	>	self.blackWhiteListParser = SEBlackWhiteListParser(self.seWhiteList, seBlackList, common.logger())
35	>
36	>	## check if has been asked for a non default file to store/read analyzed fileBlocks
37	>	defaultName = common.work_space.shareDir()+'AnalyzedBlocks.txt'
38	>	self.fileBlocks_FileName = os.path.abspath(self.cfg_params.get('CMSSW.fileblocks_file',defaultName))
39
40
41		def checkUserSettings(self):
#	Line 42 | Line 66 | class JobSplitter:
66		self.total_number_of_events = 0
67		self.selectTotalNumberEvents = 0
68
69	+	return
70	+
71	+	def checkLumiSettings(self):
72	+	"""
73	+	Check to make sure the user has specified enough information to
74	+	perform splitting by Lumis to run the job
75	+	"""
76	+	settings = 0
77	+	if self.cfg_params.has_key('CMSSW.lumis_per_job'):
78	+	self.lumisPerJob =int( self.cfg_params['CMSSW.lumis_per_job'])
79	+	self.limitJobLumis = True
80	+	settings += 1
81	+
82	+	if self.cfg_params.has_key('CMSSW.number_of_jobs'):
83	+	self.theNumberOfJobs =int( self.cfg_params['CMSSW.number_of_jobs'])
84	+	self.limitNJobs = True
85	+	settings += 1
86	+
87	+	if self.cfg_params.has_key('CMSSW.total_number_of_lumis'):
88	+	self.totalNLumis = int(self.cfg_params['CMSSW.total_number_of_lumis'])
89	+	self.limitTotalLumis = (self.totalNLumis != -1)
90	+	settings += 1
91	+
92	+	if settings != 2:
93	+	msg = 'When splitting by lumi section you must specify two and only two of:\n'
94	+	msg += '  number_of_jobs, lumis_per_job, total_number_of_lumis'
95	+	raise CrabException(msg)
96	+	if self.limitNJobs and self.limitJobLumis:
97	+	self.limitTotalLumis = True
98	+	self.totalNLumis = self.lumisPerJob * self.theNumberOfJobs
99	+
100	+	# Has the user specified runselection?
101	+	if (self.cfg_params.has_key('CMSSW.runselection')):
102	+	common.logger.info('You have specified runselection and split by lumi.')
103	+	common.logger.info('Good lumi list will be the intersection of runselection and lumimask or ADS (if any).')
104	+	return
105	+
106	+	def ComputeSubBlockSites( self, blockSites ):
107	+	"""
108	+	"""
109	+	sub_blockSites = {}
110	+	for k,v in blockSites.iteritems():
111	+	sites=self.blackWhiteListParser.checkWhiteList(v)
112	+	if sites : sub_blockSites[k]=v
113	+	if len(sub_blockSites) < 1:
114	+	msg = 'WARNING: the sites %s is not hosting any part of data.'%self.seWhiteList
115	+	raise CrabException(msg)
116	+	return sub_blockSites
117
118		########################################################################
119		def jobSplittingByEvent( self ):
#	Line 77 | Line 149 | class JobSplitter:
149		self.useParent = int(self.cfg_params.get('CMSSW.use_parent',0))
150		noBboundary = int(self.cfg_params.get('CMSSW.no_block_boundary',0))
151
152	+	if noBboundary == 1:
153	+	if self.total_number_of_events== -1:
154	+	msg = 'You are selecting no_block_boundary=1 which does not allow to set total_number_of_events=-1\n'
155	+	msg +='\tYou shoud get the number of event from DBS web interface and use it for your configuration.'
156	+	raise CrabException(msg)
157	+	if len(self.seWhiteList) == 0 or  len(self.seWhiteList.split(',')) != 1:
158	+	msg = 'You are selecting no_block_boundary=1 which requires to choose one and only one site.\n'
159	+	msg += "\tPlease set se_white_list with the site's storage element name."
160	+	raise  CrabException(msg)
161	+	blockSites = self.ComputeSubBlockSites(blockSites)
162	+
163		# ---- Handle the possible job splitting configurations ---- #
164		if (self.selectTotalNumberEvents):
165		totalEventsRequested = self.total_number_of_events
#	Line 300 | Line 383 | class JobSplitter:
383		allBlock = []
384
385		blockCounter = 0
386	+	saveFblocks =''
387		for block in blocks:
388		if block in jobsOfBlock.keys() :
389		blockCounter += 1
#	Line 310 | Line 394 | class JobSplitter:
394		if len(sites) == 0:
395		noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
396		bloskNoSite.append( blockCounter )
397	+	else:
398	+	saveFblocks += str(block)+'\n'
399	+	writeTXTfile(self, self.fileBlocks_FileName , saveFblocks)
400
401		common.logger.info(screenOutput)
402		if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
#	Line 319 | Line 406 | class JobSplitter:
406		virgola = ","
407		for block in bloskNoSite:
408		msg += ' ' + str(block) + virgola
409	<	msg += '\n               Related jobs:\n                 '
409	>	msg += '\n\t\tRelated jobs:\n                 '
410		virgola = ""
411		if len(noSiteBlock) > 1:
412		virgola = ","
413		for range_jobs in noSiteBlock:
414		msg += str(range_jobs) + virgola
415	<	msg += '\n               will not be submitted and this block of data can not be analyzed!\n'
415	>	msg += '\n\t\twill not be submitted and this block of data can not be analyzed!\n'
416		if self.cfg_params.has_key('GRID.se_white_list'):
417	<	msg += 'WARNING: SE White List: '+self.cfg_params['GRID.se_white_list']+'\n'
418	<	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
419	<	msg += 'Please check if the dataset is available at this site!)\n'
417	>	msg += '\tWARNING: SE White List: '+self.cfg_params['GRID.se_white_list']+'\n'
418	>	msg += '\t(Hint: By whitelisting you force the job to run at this particular site(s).\n'
419	>	msg += '\tPlease check if the dataset is available at this site!)'
420		if self.cfg_params.has_key('GRID.ce_white_list'):
421	<	msg += 'WARNING: CE White List: '+self.cfg_params['GRID.ce_white_list']+'\n'
422	<	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
423	<	msg += 'Please check if the dataset is available at this site!)\n'
421	>	msg += '\tWARNING: CE White List: '+self.cfg_params['GRID.ce_white_list']+'\n'
422	>	msg += '\t(Hint: By whitelisting you force the job to run at this particular site(s).\n'
423	>	msg += '\tPlease check if the dataset is available at this site!)\n'
424
425		common.logger.info(msg)
426
427		if bloskNoSite == allBlock:
428	<	raise CrabException('No jobs created')
428	>	msg += 'Requested jobs cannot be Created! \n'
429	>	if self.cfg_params.has_key('GRID.se_white_list'):
430	>	msg += '\tWARNING: SE White List: '+self.cfg_params['GRID.se_white_list']+'\n'
431	>	msg += '\t(Hint: By whitelisting you force the job to run at this particular site(s).\n'
432	>	msg += '\tPlease check if the dataset is available at this site!)'
433	>	if self.cfg_params.has_key('GRID.ce_white_list'):
434	>	msg += '\tWARNING: CE White List: '+self.cfg_params['GRID.ce_white_list']+'\n'
435	>	msg += '\t(Hint: By whitelisting you force the job to run at this particular site(s).\n'
436	>	msg += '\tPlease check if the dataset is available at this site!)\n'
437	>	raise CrabException(msg)
438
439		return
440
#	Line 347 | Line 443 | class JobSplitter:
443		def jobSplittingByRun(self):
444		"""
445		"""
350	–	from sets import Set
351	–	from WMCore.JobSplitting.RunBased import RunBased
352	–	from WMCore.DataStructs.Workflow import Workflow
353	–	from WMCore.DataStructs.File import File
354	–	from WMCore.DataStructs.Fileset import Fileset
355	–	from WMCore.DataStructs.Subscription import Subscription
356	–	from WMCore.JobSplitting.SplitterFactory import SplitterFactory
357	–	from WMCore.DataStructs.Run import Run
446
447		self.checkUserSettings()
448		blockSites = self.args['blockSites']
#	Line 393 | Line 481 | class JobSplitter:
481		jobfactory = splitter(subs)
482
483		#loop over all runs
396	–	set = Set(runList)
484		list_of_lists = []
485		jobDestination = []
486	+	list_of_blocks = []
487		count = 0
488		for jobGroup in  jobfactory():
489		if count <  self.theNumberOfJobs:
#	Line 407 | Line 495 | class JobSplitter:
495		list_of_lists.append([fullString,str(-1),str(0)])
496		#need to check single file location
497		jobDestination.append(res['locations'])
498	+	list_of_blocks.append(res['block'])
499		count +=1
500	<	# prepare dict output
500	>	# prepare dict output
501		dictOut = {}
502		dictOut['params']= ['InputFiles','MaxEvents','SkipEvents']
503		dictOut['args'] = list_of_lists
504		dictOut['jobDestination'] = jobDestination
505		dictOut['njobs']=count
506
507	+	self.cacheBlocks(list_of_blocks,jobDestination)
508	+
509		return dictOut
510
511		def getJobInfo( self,jobGroup ):
#	Line 428 | Line 519 | class JobSplitter:
519		for loc in file['locations']:
520		if tmp_check < 1 :
521		locations.append(loc)
522	+	res['block']= file['block']
523		tmp_check = tmp_check + 1
432	–	### qui va messo il check per la locations
524		res['lfns'] = lfns
525		res['locations'] = locations
526		return res
#	Line 474 | Line 565 | class JobSplitter:
565
566		managedGenerators =self.args['managedGenerators']
567		generator = self.args['generator']
568	<	firstRun = self.cfg_params.get('CMSSW.first_run',None)
568	>	firstLumi = self.cfg_params.get('CMSSW.first_lumi', 1)
569
570		self.prepareSplittingNoInput()
571
#	Line 493 | Line 584 | class JobSplitter:
584		self.list_of_args = []
585		for i in range(self.total_number_of_jobs):
586		## Since there is no input, any site is good
587	<	jobDestination.append([""]) #must be empty to write correctly the xml
587	>	jobDestination.append([""]) # must be empty to correctly write the XML
588		args=[]
589	<	if (firstRun):
590	<	## pythia first run
500	<	args.append(str(firstRun)+str(i))
589	>	if (firstLumi): # Pythia first lumi
590	>	args.append(str(int(firstLumi)+i))
591		if (generator in managedGenerators):
592		args.append(generator)
593		if (generator == 'comphep' and i == 0):
#	Line 511 | Line 601 | class JobSplitter:
601
602		dictOut = {}
603		dictOut['params'] = ['MaxEvents']
604	<	if (firstRun):
605	<	dictOut['params'] = ['FirstRun','MaxEvents']
606	<	if ( generator in managedGenerators ) :
607	<	dictOut['params'] = ['FirstRun', 'Generator', 'FirstEvent', 'MaxEvents']
604	>	if (firstLumi):
605	>	dictOut['params'] = ['FirstLumi','MaxEvents']
606	>	if (generator in managedGenerators):
607	>	dictOut['params'] = ['FirstLumi', 'Generator', 'FirstEvent', 'MaxEvents']
608		else:
609		if (generator in managedGenerators) :
610		dictOut['params'] = ['Generator', 'FirstEvent', 'MaxEvents']
#	Line 565 | Line 655 | class JobSplitter:
655
656		def jobSplittingByLumi(self):
657		"""
658	+	Split task into jobs by Lumi section paying attention to which
659	+	lumis should be run (according to the analysis dataset).
660	+	This uses WMBS job splitting which does not split files over jobs
661	+	so the job will have AT LEAST as many lumis as requested, perhaps
662	+	more
663		"""
664	<	return
664	>
665	>	common.logger.debug('Splitting by Lumi')
666	>	self.checkLumiSettings()
667	>
668	>	blockSites = self.args['blockSites']
669	>	pubdata = self.args['pubdata']
670	>
671	>	lumisPerFile  = pubdata.getLumis()
672	>
673	>	# Make the list of WMBS files for job splitter
674	>	fileList = pubdata.getListFiles()
675	>	thefiles = Fileset(name='FilesToSplit')
676	>	for jobFile in fileList:
677	>	block = jobFile['Block']['Name']
678	>	try:
679	>	jobFile['Block']['StorageElementList'].extend(blockSites[block])
680	>	except:
681	>	continue
682	>	wmbsFile = File(jobFile['LogicalFileName'])
683	>	[ wmbsFile['locations'].add(x) for x in blockSites[block] ]
684	>	wmbsFile['block'] = block
685	>	for lumi in lumisPerFile[jobFile['LogicalFileName']]:
686	>	wmbsFile.addRun(Run(lumi[0], lumi[1]))
687	>	thefiles.addFile(wmbsFile)
688	>
689	>	# Create the factory and workflow
690	>	work = Workflow()
691	>	subs = Subscription(fileset    = thefiles,    workflow = work,
692	>	split_algo = 'LumiBased', type     = "Processing")
693	>	splitter = SplitterFactory()
694	>	jobFactory = splitter(subs)
695	>
696	>	list_of_lists = []
697	>	jobDestination = []
698	>	jobCount = 0
699	>	lumisCreated = 0
700	>	list_of_blocks = []
701	>	if not self.limitJobLumis:
702	>	self.lumisPerJob = pubdata.getMaxLumis() // self.theNumberOfJobs + 1
703	>	common.logger.info('Each job will process about %s lumis.' %
704	>	self.lumisPerJob)
705	>
706	>	for jobGroup in jobFactory(lumis_per_job = self.lumisPerJob):
707	>	for job in jobGroup.jobs:
708	>	if (self.limitNJobs and jobCount >= self.theNumberOfJobs):
709	>	common.logger.info('Limit on number of jobs reached.')
710	>	break
711	>	if (self.limitTotalLumis and lumisCreated >= self.totalNLumis):
712	>	common.logger.info('Limit on number of lumis reached.')
713	>	break
714	>	lumis = []
715	>	lfns  = []
716	>	locations = []
717	>	blocks = []
718	>	firstFile = True
719	>	# Collect information from all the files
720	>	for jobFile in job.getFiles():
721	>	doFile = False
722	>	if firstFile:  # Get locations from first file in the job
723	>	for loc in jobFile['locations']:
724	>	locations.append(loc)
725	>	blocks.append(jobFile['block'])
726	>	firstFile = False
727	>	# Accumulate Lumis from all files
728	>	for lumiList in jobFile['runs']:
729	>	theRun = lumiList.run
730	>	for theLumi in list(lumiList):
731	>	if (not self.limitTotalLumis) or \
732	>	(lumisCreated < self.totalNLumis):
733	>	doFile = True
734	>	lumisCreated += 1
735	>	lumis.append( (theRun, theLumi) )
736	>	if doFile:
737	>	lfns.append(jobFile['lfn'])
738	>	fileString = ','.join(lfns)
739	>	lumiLister = LumiList(lumis = lumis)
740	>	lumiString = lumiLister.getCMSSWString()
741	>	list_of_lists.append([fileString, str(-1), str(0), lumiString])
742	>	list_of_blocks.append(blocks)
743	>	jobDestination.append(locations)
744	>	jobCount += 1
745	>	common.logger.debug('Job %s will run on %s files and %s lumis '
746	>	% (jobCount, len(lfns), len(lumis) ))
747	>
748	>	common.logger.info('%s jobs created to run on %s lumis' %
749	>	(jobCount, lumisCreated))
750	>
751	>	# Prepare dict output matching back to non-WMBS job creation
752	>	dictOut = {}
753	>	dictOut['params'] = ['InputFiles', 'MaxEvents', 'SkipEvents', 'Lumis']
754	>	dictOut['args'] = list_of_lists
755	>	dictOut['jobDestination'] = jobDestination
756	>	dictOut['njobs'] = jobCount
757	>
758	>	self.cacheBlocks(list_of_blocks,jobDestination)
759	>
760	>	return dictOut
761	>
762	>	def cacheBlocks(self, blocks,destinations):
763	>
764	>	saveFblocks=''
765	>	for i in range(len(blocks)):
766	>	sites=self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(destinations[i]))
767	>	if len(sites) != 0:
768	>	for block in blocks[i]:
769	>	saveFblocks += str(block)+'\n'
770	>	writeTXTfile(self, self.fileBlocks_FileName , saveFblocks)
771	>
772		def Algos(self):
773		"""
774		Define key splittingType matrix

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines